Method for defining threshold stress curves utilized in fatigue and damage tolerance analysis

Abstract

A method for defining a threshold stress curve utilized in fatigue and/or damage tolerance analysis, the method including conducting a direct test method on a plurality of specimens with a notch to generate raw data, the notch in each of the specimens having a notch dimension; determining a plurality of analyzed data points based upon the raw data; determining a regression line based upon the plurality of data points, the regression line including an exponential portion and a power law portion; and connecting the exponential portion and the power law portion of the regression line to form a threshold stress curve based on the regression line.

The invention claimed is: 
     
       1. A method for defining a threshold stress curve utilized in damage tolerance analysis, the method comprising:
 conducting a direct test method on a plurality of specimens with a notch to generate raw data, the notch in each of the specimens having a notch dimension, the step of conducting the direct test method comprises:
 applying a cyclic load to each of the specimens in a plurality of specimens with a notch until a first crack emanating from the notch is detected, the cyclic load having a maximum load and a minimum load; and 
 applying a subsequent cyclic load to each of the specimens until extension of the first crack to form a second crack emanating from the first crack is detected, the subsequent cyclic load having the same maximum load but a greater minimum load; 
 
 determining a plurality of analyzed data points based upon the raw data; 
 determining a regression line based upon the plurality of data points, the regression line comprising an exponential portion and a power law portion; and 
 connecting the exponential portion and the power law portion of the regression line to form a threshold stress curve based on the regression line. 
 
     
     
       2. The method according to  claim 1 , wherein each of the specimens has a square cross-sectional portion and the notch is located in a corner of the square cross-sectional portion. 
     
     
       3. The method according to  claim 1 , wherein the step of determining a plurality of analyzed data points comprises:
 categorizing the raw data into crack data or flaw data. 
 
     
     
       4. The method according to  claim 3 , wherein the step of determining a plurality of analyzed data points comprises:
 selecting the categorized raw data of a substantially similar dimension. 
 
     
     
       5. The method according to  claim 4 , wherein the step of determining a plurality of analyzed data points comprises:
 collapsing at least a portion of the categorized raw data to a nominal dimension. 
 
     
     
       6. The method according to  claim 4 , wherein the substantially similar dimension comprises at least one of the following: a flaw size, a notch length, and a crack length. 
     
     
       7. The method according to  claim 4 , wherein the step of adjusting the raw data comprises:
 generating the plurality of analyzed data points from the categorized and selected raw data. 
 
     
     
       8. The method according to  claim 7 , wherein the step of determining a plurality of analyzed data points comprises:
 plotting the plurality of analyzed data points onto an X-Y coordinate plane. 
 
     
     
       9. The method according to  claim 7 , wherein the step of determining a regression line based upon the plurality of data points further comprising:
 identifying a stress ratio (R) based on the raw data, the stress ratio comprising at least a negative stress ratio, a zero stress ratio, and at least a positive stress ratio. 
 
     
     
       10. The method according to  claim 9 , wherein the step of determining a regression line based upon the plurality of data points further comprising:
 if a portion of the plurality of analyzed data points has at least a negative stress ratio (R−), performing a regression with an exponential fitting of the plurality of analyzed data points having the negative stress ratio to generate the exponential portion of the regression line; and 
 if a portion of the plurality of the analyzed data points has at least a positive stress ratio (R+), performing a regression with a power law fitting of the plurality of analyzed data points having the positive stress ratio to generate the power law portion of the regression line. 
 
     
     
       11. The method according to  claim 9 , wherein the step of connecting comprises:
 performing a derivative function to have a smooth connection between the exponential portion and the power law portion of the regression line. 
 
     
     
       12. The method according to  claim 1 , wherein points below the threshold stress curve indicate that a flaw or crack is not growing and points above the threshold stress curve indicate that a flaw or crack is growing.